ZMIME
Google · Released Dec 17, 2025

Gemini 3 Flash Preview

New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs.

  • Proprietary
  • Reasoning
  • Tool calling
  • Vision
  • Audio input
  • Video input
Capability
151.8ECI · #43 of 148
Input
$0.50per 1M tokens
Output
$3.00per 1M tokens
Context
1.05M66K max output
Pricing

What it costs

Input
$0.50per million tokens
Output
$3.00per million tokens
Cached input
$0.05per million tokens
Blended (3:1)
$1.13pricier than 58% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$2.50
Team assistant25M input + 5M output tokens$27.50
Production app250M input + 50M output tokens$275.00

Official Google API price, as listed on models.dev.

Capability

Independent benchmarks

151.8Epoch Capabilities Index
#43 of 148 scored models
88Median 146167

The shaded band is Epoch AI’s confidence range (150.4–153.7); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions · high setting89.4%
  • FrontierMath Tiers 1–3Research-level mathematics51.2%
  • OTIS Mock AIME 2024–2025Competition mathematics · high setting95.6%
  • SWE-bench VerifiedFixing real GitHub issues75.4%
  • SimpleQA VerifiedShort factual questions · high setting66.8%

Scores from Epoch AI, run independently of Google.

Specifications

The details

Lab
Google
Released
Dec 17, 2025
Knowledge cutoff
Jan 2025
Context window
1,048,576 tokens
Max output
65,536 tokens
Inputs
Text, Images, PDFs, Audio, Video
Output
Text
Reasoning
Adjustable effort minimal · low · medium · high
Tool calling
Yes
Structured output
Yes
Weights
Proprietary
API model ID
gemini-3-flash-previewon Google
Availability
20 API providerslisted on models.dev

Google model documentation

Lab-reported

What Google claims

Published by the lab at launch. Settings vary, so compare these only with care.

BenchmarkScoreSettingSource
SWE-Bench Pro34.63 resolve rate—Source
SWE-Atlas Codebase QnA8.2—Source
SWE-Atlas Refactoring10—Source
SWE-Atlas Test Writing30.3—Source
Same lab

More from Google

Questions

About Gemini 3 Flash Preview

How much does Gemini 3 Flash Preview cost?

Gemini 3 Flash Preview costs $0.50 input / $3.00 output per million tokens (official Google API price). Cached input is $0.05 per million tokens. At a 3:1 input-to-output mix that is $1.13 per million tokens, more expensive than 58% of the 360 priced models we track.

What is the context window of Gemini 3 Flash Preview?

Gemini 3 Flash Preview accepts up to 1,048,576 tokens per request and can write up to 65,536 tokens in one response.

How good is Gemini 3 Flash Preview?

Epoch AI gives Gemini 3 Flash Preview a Capabilities Index score of 151.8 (likely range 150.4–153.7), ranking it #43 of 148 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 89.4%, FrontierMath Tiers 1–3 51.2%, OTIS Mock AIME 2024–2025 95.6%, SWE-bench Verified 75.4%, SimpleQA Verified 66.8%.

Is Gemini 3 Flash Preview open source?

No. Gemini 3 Flash Preview is proprietary; you use it through Google’s API or partner platforms.

What inputs does Gemini 3 Flash Preview support?

Gemini 3 Flash Preview accepts text, images, PDFs, audio and video and replies in text. It is a reasoning model with minimal, low, medium and high effort settings, supports tool calling and can return structured JSON output.

When was Gemini 3 Flash Preview released?

Google released Gemini 3 Flash Preview on Dec 17, 2025. Its training data runs to Jan 2025.

What are the best alternatives to Gemini 3 Flash Preview?

The closest current models from other labs on capability, price and release date are Grok 4.20 (Reasoning) (xAI, ECI 152.0, $1.25 / $2.50), Kimi K2.5 (Moonshot AI, ECI 148.0, $0.60 / $3.00), Qwen3.6 Plus (Alibaba (Qwen), ECI 147.6, $0.50 / $3.00) and GLM-5.2 (Z.ai (Zhipu), ECI 151.8, $1.40 / $4.40).